2020/11/10 by Jonatas B. C. Chagas, Markus Wagner, Chagas, Jonatas B. C. +1
Engineering · #Advanced Manufacturing and Logistics Optimization #FOS: Computer and information sciences #Neural and Evolutionary Computing (cs.NE) #Optimization and Packing Problems #Vehicle Routing Optimization Methods
paper · pdf · doi:10.48550/arxiv.2011.05081
openalex publication_date 2020/11/10 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
Many real-world optimization problems have multiple interacting components.\nEach of these can be NP-hard and they can be in conflict with each other, i.e.,\nthe optimal solution for one component does not necessarily represent an\noptimal solution for the other components. This can be a challenge for\nsingle-objective formulations, where the respective influence that each\ncomponent has on the overall solution quality can vary from instance to\ninstance. In this paper, we study a bi-objective formulation of the traveling\nthief problem, which has as components the traveling salesperson problem and\nthe knapsack problem. We present a weighted-sum method that makes use of\nrandomized versions of existing heuristics, that outperforms participants on 6\nof 9 instances of recent competitions, and that has found new best solutions to\n379 single-objective problem instances.\n